Canon lens twist trick for a Nikon lens?

I`m looking for a short tele (60-200, zoom or prime) to fit my D5100, so aps-c, for time lapse work. My Canon cameras have the neat lens twist trick where you hold down the depth of field preview lever while untwisting the lens. It locks the set aperture , which eliminates aperture flicker in time lapse. The few modern (af) Nikon lenses I have owned have all stopped down to a useless f22 when untwisting.

Are there any lenses for Nikon aps-c in the range I mentioned which don`t have this behaviour? (I`d consider third party lenses).

My old 28-105mm (film) works but is very soft.

I don`t want to work at open aperture btw.

Thanks for any suggestions.

Ian

As I understand it, the problem is that auto aperture lenses supposedly do not always stop down to precisely the same aperture every time, even in manual or aperture priority modes, a limitation that only becomes apparent in time lapse work. If this is the case, I don't know why the 28-105 would behave any differently when 'untwisted' - have you tried other Nikon lenses with aperture rings? Incidentally, I don't find this lens soft when stopped down a bit. One thing you might try is a Leitax conversion, which will allow you to use a Leica R, Contax/Yashica or Olympus OM lens on Nikon: http://www.leitax.com/ Converted lenses have entirely manual aperture control.

Perhaps I have a lemon Richard but my copy is soft.

I think I may buy the 55-200mm and jam the aperture at about f8 with a piece of plastic. I don`t use the camera for anything but time lapse anyway.

E lenses and lenses with aperture ring allow you to have precisely reproducible exposures it is basically only the G lenses which have this issue. You need to allow the use of the aperture ring in Ai P, AF, AF D lenses to take advantage of it by using a custom function on the camera. I don't think G lenses can be fixed to a constant aperture.

I don't think G lenses can be fixed to a constant aperture.

Same thoughts here - since the linkage is purely mechanical, as long as the lever in the camera can engage the actuator in the lens, the lens will stop down when the shutter is released and open again afterward. If twisted enough that the lever no longer engages, a G lens will have its aperture close down fully. It appears Ian is aware of the trick to "jam" the actuator in the lens with a piece of plastic; I read about it but have not tried it myself.

I`ve read that shooting time lapse in live view locks the aperture. Is that correct?

I can`t use the D5100`s internal intervalometer when in live view (greyed out) but I can fire it with an external intervalometer when in live view.

It`s obviously more draining on the battery but it`d be worth it to me.

Any manual focus Ai or Ai-s Nikkor will allow setting a fixed aperture from its barrel ring. I imagine you don't need AF for time-lapse.<p>

Most of Nikon's MF primes are plenty sharp enough for video resolution, especially stopped down a couple of stops.<p><br>

The same goes for any of the older AF or AF-D lenses that have an aperture ring.

I`ve read that shooting time lapse in live view locks the aperture. Is that correct?

Yes and no. The aperture will open after a shot was taken, then stop down again; it doesn't stay closes all the time. Not sure why it opens in between, it's something that probably could be changed with through firmware (if Nikon was so inclined).

 

allow setting a fixed aperture from its barrel ring

Setting, yes, but the aperture still opens after each shot and needs to stop down for the next. Whether or not that is more reproducible with lenses that have a aperture ring vs those that don't (G) is something I don't know. G lenses certainly don't reproduce the same opening with each shot and I doubt that the lenses with an aperture ring fare better. I have on leitax-modified lens and on that one I can be sure that the aperture is the same every time as it is set and then stays there until another setting is chosen.

Dieter, the mechanical stop-down is extremely reproducible, within less than 1/10th of a stop I'd estimate, and well within what could be expected of lighting variations between time-lapse shots. It's only photography. Extreme accuracy has never been needed when a one-stop change means a doubling or halving of light!

Most timelapse software, such as LRTimelapse, has the ability to adjust the exposure of individual 'errant' frames to prevent exposure variation induced 'flicker'. Edited February 27, 2017 by mike_halliwell

the mechanical stop-down is extremely reproducible, within less than 1/10th of a stop I'd estimate

Well, it is a issue with G lenses for sure, and unless lenses with an aperture ring somehow function differently on stopping down, it can be expected to be an issue with them as well. Saw a time lapse video of aperture closing yesterday and the shape and/or size of the aperture varied a bit; I just don't know if that variation is sufficient to account for the flicker that is clearly often an issue with time lapse photography. If it wasn't, then the solution Mike mentions wasn't needed at all. Though, of course, the aperture is only one part of the exposure equation, the shutter speed is the other and there's some slit variation to be expected as well.

unless lenses with an aperture ring somehow function differently on stopping down,

 

If you're using the aperture ring (as opposed to setting the aperture to its minimum and letting the camera control it), they do.<br />

<br />

On a G lens, or when the camera is otherwise controlling the aperture on an aperture ring lens, the position of the aperture lever in the camera controls how much the aperture opens. This doesn't have a lot of travel, and has to move a floppy mechanical aperture lever that potentially goes some distance into the lens (possibly through a teleconverter). It has to do this with enough torque to stop the movement of the aperture against the spring pressure, while being fast enough to keep up with frame rate. I'm not surprised the mechanism is slightly variable. Although I'd still love to know why nothing recent can control the aperture of an AI-S lens this way. (For an AI lens, the movement of the lever isn't linear against aperture, meaning setting the aperture is infeasible.) Incidentally, I'm not actually sure how the aperture lever position movement is implemented in the camera, other than that some cameras seem to have a separate motor for some of it. The slowness with which my D810 changed aperture in live view (when pressing DoF preview on my 35mm f/1.4 Sigma) makes me think what's going on is complicated. I'm sure it's highly engineered, but given what it has to do, I'm amazed it works at all.<br />

<br />

If you're using the aperture ring, the aperture lever just opens fully (from its position where it was holding the aperture open against spring pressure). The aperture opens under its own springs, and whacks into a hard mechanical stop moved by the aperture ring. That has the potential to be more accurately controlled than the camera-side aperture lever, and the lever just has to flip to the "fully open" position without "catching" the aperture on the lens.<br />

<br />

E (aperture) lenses can move just the aperture, electronically, without worrying about a mechanical linkage. Maybe it's because I started with Canon, but this always seemed much more sensible to me. There's still a chance of slight inaccuracy, on the basis that there still no hard mechanical stop in the system, but there's a lot less potential for slop than with a G lens.</p>

I think the software is supposed to correct anomalous exposure variation, whether it be a single passing cloud, a smallish variation in aperture stop-down or less likely variation in shutter speed.

 

 

 

I guess it might actually be quite hard to pin down exactly what causes the difference in exposure as EXIF data won't really help here.... it displays 'theoretical' values.

I think the software is supposed to correct anomalous exposure variation, whether it be a single passing cloud, a smallish variation in aperture stop-down or less likely variation in shutter speed.<br />

<br />

I guess it might actually be quite hard to pin down exactly what causes the difference in exposure as EXIF data won't really help here.... it displays 'theoretical' values.

 

If you wanted constant exposure, it shouldn't be hard to override. I had a bit of a fight with AfterEffects trying to compensate for clouds and me blocking the lights in a recent video, but I vaguely got there in the end - and did better a year ago with the same approach. Fixing the white balance was harder. I actually find this kind of thing easier to do myself: I once captured a 200fps sequence with an engineering camera (at around the turn of the millennium) which gave me a load of stills, whose brightness varied as the fluorescent lights flickered; I just wrote some code to detect the average image brightness and adjust accordingly. I could have done the same by picking an area of the image that stayed constant. White balance would have been as easy. I'm sure you can do it in video processing software, but as usual I struggle with Adobe's interfaces. I found this the biggest limitation with the high frame rate on the 1V: I either have to do LED lighting (for which I now have options, fortunately) or shoot outside in daylight, because mains-powered lights flicker too much. I look forward to the timed anti-flicker option from the 7D2 and D500 making it to the D810 successor.

Someone told me the other day that 'dimmable' LED studio lights use PWM to achieve there variation. I guess that means they're flashing/pulsing/flickering too...just really fast? Edited February 27, 2017 by mike_halliwell

Someone told me the other day that 'dimmable' LED studio lights use PWM to achieve there variation. I guess that means they're flashing too...just really fast?

 

They probably do. Most high-end flashlights do too, but usually fast enough not to cause a problem. Candlepowerforums and flashlightreviews.ca test for this kind of thing. I do have a "thrower" that detectably buzzes on low power settings, but my big flooders are much better behaved than my kitchen spotlights. And brighter. :-)

mm there are many non-nikon lenses which are "Fully Manual" like petzvall's and LensBaby's , maybe one of those those would answer to your challenge ?

 

Examle : https://www.dpreview.com/articles/6353634878/lensbaby-edge-80-quick-review

 

Or maybe something like https: //petapixel.com/2016/03/10/laowa-105mm-f2-lens-uses-special-element-crazy-smooth-bokeh/

Which utilizes 2 apertutres , both fully manual...

Edited February 27, 2017 by c.p.m._van_het_kaar

I've noticed something regarding the E lenses which is somewhat problematic. In very cold temperatures (e.g. -20C) and high humidity which we encounter during the coldest weeks in Finland, when photographing near open water (e.g. before the sea freezes over), the camera can fail to correctly control the aperture after about 1 hour of use (it typically works fine for a while). The symptoms can be the camera simply won't respond to controls, the aperture in live view won't stop down, and actual images taken can be overexposed as the stopping down does not take place correctly. I've seen problems in these conditions with the D700, D800 and D810, and I initially assumed it was a camera-related problem, but then I realized most of these experiences had taken place with E lenses. I had one case of overexposure with a G lens that occurred twice but it was due to some dirt that had gotten inside the aperture control mechanism of the 70-200/2.8G (1st version). This was fixed in repair. However, the stop-down problems I've had with E lenses seem persistent in these winter conditions and occur on a regular basis. Once I figured out it was probably due to the electrically controlled aperture, I realized there is a simple solution to the problem: take off the lens and put it back. I believe the humid air freezes on the contacts between the lens and body, and prevent the aperture control from happening correctly. I realize that the mechanical aperture stop down mechanism of G and earlier lenses can also have this kind of issues if they're exposed to dirt, but still, in the kind of conditions that I shoot in, I've had a much higher frequency of failure of E lenses to stop down correctly. The problem has exhibited itself with the 24, 45, and 85 PC-E, 70-200/2.8E at least, maybe some other lenses but mostly those. I have some Canon using friends, some of whom never seem to have problems with the camera operation in the cold, and others who report occasional problems with camera malfunction in the cold, but I've not heard of them having this particular problem with the electromagnetically controlled aperture. Perhaps the difference is that Canon's lens pins point towards the body whereas Nikon's point outwards, towards the bayonet, and maybe they are a bit more exposed? I have not had other communications problems with Nikon cameras or lenses that I can recall.

 

Anyway, since I now know that the problem can be rectified by taking off the lens and putting it back on, I'm confident that I can work around the problem. I just felt prudent to mention that perfect operation is not guaranteed with the E system and the G system has its own advantages. G lenses may vary from lens type to another as well, as with some lenses the distance between the lens mount and the physical aperture is quite long, leading to longer mechanical system which must function precisely and quickly, whereas with some other lens types the aperture is located right next to the mount, leading to an easier implementation and probably less variation from exposure to exposure.

Ilkka, do they all overexpose? A kinda lube-thickening 'sticky' aperture blade issue?

 

Maybe the lubricant in E-lens aperture mechanisms is different?

 

 

If one fired a burst of, say, 10 frames, I'd expect the blade friction to lessen throughout the burst as the lube warms....maybe?

 

If it's an actual communication issue, I wouldn't expect the lens to get the wrong aperture as such as to not function at all.

I don't think it's about the lubing of the aperture as unmounting the lens and putting it back on and firing a shot solves the issue and the lens works perfectly after that ... for a while at least. (The first shot after the unmount/mount cycle is needed to set the aperture setting to its correct value; it may not come out correctly, but after that it seems to work as if nothing had been wrong.)

 

The camera does give error messages when this problem occurs, and show the wrong aperture when in live view, and refuse to take the picture, but sometimes it will let a picture to be taken but the exposure is often wrong.

If you're using the aperture ring (as opposed to setting the aperture to its minimum and letting the camera control it), they do.

What you wrote is what I assumed is the way it works, I just wasn't sure. But as you say "the potential to be more accurate"; only testing can reveal if it is indeed the case.<br><br> I am quite certain there are plenty of posts on "aperture flicker" during time lapse on the internet and what possible solutions there are. Certainly appears to be an issue and the solution when using Nikon cameras is not found in partially unmounting the lens.

It's only how I assume it works as well, Dieter, but it's a fairly educated guess. :-) Digital fixing ought to be simple enough that it's a non-issue. I'd not be surprised if timing relative to interior lights causes as many problems as anything. The "use an old lens with no aperture coupling" isn't the worst idea in the world, and, so long as it doesn't land on the sensor, nor is the "wedge something in the aperture lever" idea - but they're certainly not ideal.</p>

Chasseur d'Images used to run tests of exposure reproducibility as part of the camera tests and it was found the A and M modes in Nikon film cameras had less variability than S and P modes in their lab tests. I can only guess one explanation: in using the A and M modes, the lens aperture ring was used to set the aperture, bypassing the reproducibility of the camera/lens mechanical coupling. However, this difference was smaller in top level cameras (F5) if I recall the details correctly. This was in late 1990s. I don't know if the reproducibility of body controlled aperture has improved. I guess it is possible. Anyway Nikon themselves advertise the improved reproducibility of exposure using E lenses at high fps.

Regarding AF lenses with an aperture ring. I don't think you can use the aperture ring on a D5xxx camera without a fudge. I know you can't on a D3xxx. You'll get an error unless the aperture ring is locked, and there's no override as there is on the D7xxx. The only way to trick your way around this is to mechanically wedge the minimum aperture switch on the body. I have not played with this enough to determine how the meter functions then, but I think you could use the lens in manual mode this way.

"Aperture flicker" might easily be confused with inconsistent lighting.<p>

It's hard to imagine an (amateur) lighting setup that wouldn't be subject to the vaguaries of battery or mains voltage variation. Batteries change voltage with temperature and time, as does the regulation of their associated electronics.<p>

For example: Speedlight circuitry is designed to cycle and top up its capacitor charge when the voltage falls below a preset level. So it's quite possible to catch a speedlight at the ebb and flow of its capacitor voltage points. (Capacitor energy varies as the square of its voltage.)<p>Natural lighting has hardly a hope of being constant from shot to shot, and mains-powered lighting can easily vary by 10% unless additional voltage regulation is included in the supply. Then there's the 100 or 120 Hertz flicker of hotlights or fluorescents, which can easily strobe with a high shutter speed.<p><br>The list of possible causes of exposure variation is pretty long. So pointing the finger solely at lens aperture is quite short-sighted.<p><br>

I guess the tedious answer is to balance the "exposure" of each frame in post.

Edited February 28, 2017 by rodeo_joe|1